Share your thoughts, 1 month free Claude Pro on usSee more
WorkDL logo mark

Multi-Hop QA on HotpotQA (pass@1, Avg Rank)

33.5pass@1

DAPO

10.116.17522.2528.325Apr 9, 2026
Updated 8d ago

Evaluation Results

MethodLinks
2026.04
33.53
2026.04
33.51.43
2026.04
235.29
2026.04
21.52.43
2026.04
143.57
114.57